118 



R. A. GRAHAM 



plus grande largeur pres de la base, a marges fort convexes en cet endroit, convergeant 

 ensuite suivant une ligne d= droite pour former au sommet une pointe aigue ou un peu 

 acuminee, arrondies ou tres hrievement attenueesd la base, vertes et i poilues sur les deux 

 faces . . . serrature consistant en dents assez regulieres, nombreuses, formant des 

 triangles hauts de 1-1-5 mm., droits interieurement, presque tous nettement dentes 

 exterieurement, a sommets tres aigus, separes par une distance de 3-4 mm. Capitules 

 it velus . . . Cette variete est fort voisine de la var. inciso-serrata dont elle differe 

 surtout par sa serrature double. 



Briquet regarded this variety as a near ally of var. inciso-serrata, the chief distinction 

 from the latter being the small, intermediate denticles on the exterior surface of the serra- 

 tures. The Oxford specimen, identified by Briquet, agrees with the description to a 

 certain extent, but the double serration occurs only on the lower leaves, while on the 

 upper the serratures are fewer (up to 4), and, apart from the question of immaturity, show 

 no clear indication that double serration will develop. Frankly, I can see extremely little 

 in this variety, double serration being a not uncommon feature of many forms of water 

 mint. 



(s) var. riparia (Schreber ex Schweigger & Korte) H. Braun (1890, 422). 



Stengel unterwdrts wenig behaart oder kahl, oberwdrts mit weisslichen zuriick- 

 gekriimmten Haaren ± bedeckt . . . Blatter Idnglich-lanzettlich . . . spitz aber 

 kurz gesdgt, beiderseits zerstreut, an den Hauptnerven etwas mehr mit Haaren bekleidet 

 ... an der Basis zum Blattstiele kurz zugeschweift . . . BlUthenquirle meist kleiner 

 als bei M. aquatica L. 



This is apparently a subglabrous form, with long-lanceolate, basally shortly -narrowed 

 leaves, with short, sharp serratures; and with small capitula. It would seem to be nearly 

 allied to var. ortmanniana (above), differing in having much larger leaves. Possibly one 

 is a luxuriant form of the other. This variety is included in this account as it is mentioned 

 frequently in Continental literature, although not previously listed as a British form. Many 

 herbarium specimens can be classed as this variety. 



(t) var. acuta H. Braun (1890, 422). 



M. acuta Opiz (1852, 64), sine descr. 



Stengel oberwdrts dicht behaart, Bldtter ziemlich lang gestielt. Blatter vor der 

 Mitte der Lamina an in eine lange Spitze ausgezogen, BlUthenquirle 2-3 cm. gross. 



This variety is characterised by a long, drawn-out, acute, leaf apex, and evidently has 

 large capitula. Braun regarded it as a minor variant of var. riparia, from which it differs 

 in these two characters, and in having hairier leaves and stem. Specimens classifiable as 

 var. acuta will fairly often be found, but there is every variation between a long, drawn- 

 out, acute apex to the leaf and a short, blunt one. 



The following three varieties, hitherto classedas water mints by Fraser and other authorities, 

 are rejected as being, in my opinion, of doubtful status in this group. 



(u) var. subglabra Baker (1865, 248). 



The only reference to a var. subglabra attributable to Baker appears in the above-cited 

 essay " On the English Mints." But here it is unquestionably listed as a variety of 

 M. sativa (M. X verticillata L.), nor is there any mention of a var. subglabra in the section 

 dealing with water mints. A reputedly authentic specimen at Kew, bearing a label 

 perhaps in Baker's hand, is a verticillate - not a capitate - mint, in fact M. X gentilis L. 



